The efficacy of nucleoside analog (NA) prodrugs in anticancer and antiviral treatments is often impaired by the slow intracellular activation of these drugs. NAs are administered as the uncharged nucleoside to achieve effective intracellular concentrations. Once inside the cell, NAs require conversion to their triphosphorylated form to become pharmacologically active. The sequential three-step phosphorylation is carried out by human cellular kinases, with the first step, catalyzed by nucleoside kinases, most often being rate limiting. For example, both AZT and d4T require conversion into their monophosphate form by human thymidine kinase, which for d4T is the rate-limiting step in its activation. To address the problem of inefficient NA activation we propose to structurally and kinetically characterize human thymidine kinases. Vertebrates have two different thymidine kinases, a cytosolic form (hTK1) and a mitochondrial form (hTK2). Activity of hTK1 is high in proliferating cells and peaks during the S-phase of the cell cycle, while hTK2 is constitutively expressed at low levels. To understand the molecular determinants of substrate specificity, enzymatic mechanism, and regulatory mechanisms of human thymidine kinases we propose to: (1) Solve the crystal structures of hTK1 in complex with substrates and NA prodrugs. (2) Delineate the regulatory mechanisms controlling hTK1 activity. (3) Characterize the determinants of substrate specificity and investigate the catalytic mechanism of hTK2. (4) Delineate the structural reasons that confer high d4T-phosphorylating activity of the homologous T. maritima thymidine kinase. (5) Based on the results from the previous aims, design and produce mutants of the human enzymes that possess higher activity and improved specificity for clinically relevant prodrugs. Together, the proposed work will dramatically increase our understanding of these important enzymes. In addition, the results will improve our abilities to develop NAs that are selectively activated by TK1 and not by TK2, a major goal since activation by TK2 often results in toxic side effects. A long-term goal of this project is to engineer enzymes that possess improved prodrug activation efficiencies to be employed in antiviral and anticancer therapies.